Efforts have increasingly been made for some time now to allow voice-oriented access, as well as screen-oriented access, to information on data media. A typical scenario for such access is the visual display of web content, which can be navigated with voice commands.
The linking of voice-oriented and screen-oriented input and output requires suitable processing and conversion of the transported information in addition to the corresponding input and output devices, e.g. keyboard, screen, microphone and speakers. Processing voice information above all requires significant expenditure. Functions, such as conversion from text to voice and voice recognition, take up a large proportion of the resources required in respect of memory capacity and computing power. This is why resource-intensive functions are usually only implemented on terminals in a limited form.